Showing posts with label Tononi. Show all posts
Showing posts with label Tononi. Show all posts

Sunday, October 22, 2017

Blade Runner 2049





In keeping with the previous two posts - I did get out to see Blade Runner 2049 last Saturday.  It was clearly a first rate science fiction film and I guess some viewers not used to the genre might also call it a thriller.  Visually I thought it was less stunning that the first due to the lack of street level scenes and the hectic activity on the street.  It has critical acclaim but because of the high cost is being described by some critics as a "box office bomb".  In this film replicants (bioengineered androids) have become Blade Runners.  In some reviews of the film they are referred to as bioengineered humans and that is not a trivial difference since the main plot theme is whether or not the androids can reproduce.  The focus is on K (Ryan Gosling) who is the main protagonist.   We seem him interacting with and dispatching another replicant in the initial scene.  That replicant asks for mercy on the basis that they "are the same kind" and that there is a higher calling based on the miracle that he has witnessed.  When K returns to the station (LAPD) he undergoes a rapid debriefing protocol, test questions with monitoring of various anthropometric and physiological parameters.  The meaning of the test questions is not clear but the implication is that it determines if he has stayed at his baseline or his status had been perturbed in some way.   The test is also being administered for a very different reason than the Voight-Kampff protocol since the test subject is a known replicant.

There are three generations of replicants in the film starting with K - a Nexus 9 series, to the Nexus 8 replicant he retires in the original scene, the the Nexus 7 series that dates back to Rachael in the original Blade Runner film.  Over the course of that time frame the replicant population has become less subservient and more interested in equality or autonomy.  There is a rebellious faction.  We learn later in the film based on a series of events that the common "miracle" that the replicant population refers to is the birth of a child by Rachael in the original film.  In that film in the final scene she was leaving with Deckard (Harrison Ford).  There were implications that Rachael was a specially modified replicant and in retrospect the question is whether she was modified to reproduce.   

The competing forces in the film were threefold.  First, the LAPD is invoked as the police force determined to suppress any replicant rebellion.  K is a detective for the LAPD and after discovering Rachael's remains buried at the site where he encounters the initial replicant and there is evidence that she gave birth to a child..  Second, Tyrell corporation has been replaced by the potentially more evil Wallace Corporation header by Niander Wallace.  Wallace is very explicit about the need for replicant reproduction since he does not believe that manufacturing capacity can ever meet the need for replicants in service of his corporation and its off world needs.  And finally there is the role of K as a free agent in all of this.  Does he do the bidding of his boss at LAPD or not?  His boss emphasizes the importance of killing any story that replicants have reproduced - she sees it as a game changer for civilization as they know it.  She assigns him to find and kill the child.  He is later assigned to kill Deckard for the same reason.

I will leave the plot specifics to the various reviews and descriptions already out there and concentrate on the main issues that have to do with consciousness in the film.  At one point K is asked about childhood memories and recalls being bullied by a group of boys who wanted a small hand carved horse that he was carrying.  We see him escaping the boys and burying the toy in a pile of ashes in the bottom of an old furnace.  Later he consults with an expert to determine if the memory is real or not.  She confirms that it is a real memory and that leads him to believe he may be the child of Deckard and Rachael.  I asked myself at that point if K's interest in the memory was even possible if he was a replicant.  By definition in Tononi Koch theory, this experience requires consciousness and even perfectly engineered system mimicking the human brain could not generate the human experience associated with the memory much less the integrated emotions associated with this scene.  When K finally finds Deckard he is in a state of emotional turmoil related to information that Deckard provides him about his origins.  In a shootout Deckard is captured by Wallace Corp and is in the process of being tortured to find out information about the location of his and Rachael's child.  He is both rescued by K and united with his child by K.  In both Blade Runner movies Deckard is rescued in the end by a replicant.

My summary may not match up well with other reviews about specifics.  I did not view the protocol being given to K as the Voight-Kampff protocol, since it did not seem like it was an updated version.  Keeping Tononi Koch theory in mind it would be totally unnecessary even if he was really a highly sophisticated bioengineered replicant.  It would only be necessary to place a transcranial magnetic stimulation (TMS) coil close to his brain and observe the high density electroencephalogram (EEG) pattern.  If consciousness exists the theory predicts a pattern of widespread activation and deactivation.  It should also be possible to observe the characteristic sleep EEG pattern of transitioning from consciousness to unconscious dreamless sleep and back.  Of course these androids would need to be flawlessly engineered to protect circuitry from magnetic and electrical fields that occur with these measurements. 

In summary, I thought that Blade Runner 2049 was an excellent film just based on the plot and artistry.  I can always see the distinction between real science and science fiction.  If Tononi Koch theory is accurate, it is hard to imagine that a replicant would not be obvious to conscious humans.  I guess we will need to either wait until that day comes or until the theory has more widespread acceptance and proof.  The other parallel aspect of this film is bioengineered human reproduction.  It is difficult to see how that could ever be done, especially through human sexual contact with machines.  Sexual contact with bioengineered androids is a more frequent science fiction theme these days than in the past.  It is probably easier to see how that might happen from the human side.

There is currently not enough information about human sexual consciousness to imagine how it could be built or programmed into an android.     


George Dawson, MD, DFAPA               



Thursday, October 19, 2017

Tononi Koch Test for Machine Consciousness

































In follow up to my previous post and before I saw Blade Runner 2049, I wanted to post a more modern take on the Turing Test based on a coherent theory of consciousness  by Tononi and Koch - both experts in the neuroscience of consciousness.  Their theory is the Integrated information Theory (IIT) of consciousness.  I have included the reference (1) and a graphic from their public access paper on the theory and there are also several very useful videos available to listen to the verbal descriptions of the theory.  I have been following consciousness research for at least the past 20 years including the two main listservs on this topic until they shut them down.  When a topic is so specialized, barring any breakthroughs the arguments become repetitive and a lot of time is spent bringing novices up to speed.  The videos fill a useful gap that these listservs previously addressed although I must admit  that I am always biased toward the written rather than the spoken word because it is a much more efficient information transfer for me.  The videos listed at the bottom of this page also serve another useful purpose.  The viewer is able to see how researchers in this area define consciousness and describe their theories.  I think that it is possible to notice that some of the definitions and descriptions are so vague as to have limited utility.

That is one of the reasons that I like the approach by Koch and Tonini.  I will also also say from the outset that I am not sure whether they view the theory as a joint venture or not.  As an example of what I mean looking at this specific search on consciousness finds that Tononi has been working in this area for at least 20 years.  A similar search on Koch goes back even 8 years earlier.  I don't know either of the authors but based on reading this paper it seems like a joint effort and that seems to come across in  the available videos of their presentations. (see addendum).

In the paper, that authors outline phenomenological definitions that are more exacting than any that I have seen in the past from other authors.  They are also neuroscience based and that makes a difference to me.  In various venues people often faintly praise but then lament psychiatry's emphasis on biology.  That is obviously not true or at least without reason and it also illustrates the lack of research that people do when it comes to critiquing psychiatry.  Psychiatrists have actively researched practically all forms of social, psychological, and biological etiologies of mental illness since the specialty was founded.  Any cursory review of a general psychiatric text illustrates that point.  So if a psychiatrist is focused on brain biology, it is certainly not without reason.  I previously posted a breakfast that I had with a mentor and after a long career as a psychiatrist he summed it up the way a lot of psychiatrists do: "It is all about the biology."  Critics take that to mean some kind of medical intervention.  They are certainly studied, but every other non-medical intervention has been studied as well.  It is common to read about non-medical interventions (psychotherapy, meditation, etc) altering the brain in some way.  In psychiatry that has been known within the field for at least 70 years.

There are two levels to study the work of Tononi and Koch.  The first is at the purely descriptive level.  That is the level that you will find in the first reference.  The second level is at the level of neuroscience and mathematical theory.  The authors have produced this work as well and reference it in this paper, but for the purpose of this post I am going to stay at the descriptive level and possibly post a more technical article on the advanced theory at a later date.  I will add that there are several competing theories of consciousness that I am not going to mention here.  I have studied several of them and think that they have less to offer than the Integrated information Theory (IIT) of consciousness.  I am admittedly a reductionist seeking to close the explanatory gap between brain biology and how conscious states are generated.  In some of the videos available online where there are panel discussions it is clear that the proponents of the other theories think that their own theories are correct and IIT is wrong. I have been down the rabbit hole with a few of those theories and don't want to take time to criticize them.  Feel free to look them up and form your own opinion.  For now I will focus on IIT.   

If you have never heard of Tononi, Koch, or IIT the first task is to read the paper.  I found it to be very clear in terms of definitions, postulates, and a clearly stated theory.  They point out that every experience will have an associate neural correlate of consciousness (NCC). There is currently an explanatory gap at the level of how conscious experiences are actually produced by the NCC.  They discuss the axioms necessary for a coherent phenomenology of consciousness.  From there they move on to the postulates.  Eventually they discuss how a conceptual structure that is maximally irreducible conceptual structure occurs in the brain.  These states are also known as quale.

They give a couple of examples about how conscious states occur within their theory.  They provide and example of how to calculate the quality and consciousness given a particular state containing elements (Figure 4).  They provide a clear example of the physical substrate of experience (complex), and a set of maximally irreducible cause-effect repertoires (concept), and a maximally irreducible "cause -effect structure in cause-effect space made of concepts..." or conceptual structure (quale)(p. 12).  The quantity of experience or consciousness is specified as Φmax.  The quality of experience is the form or shape of the conceptual structure. Distinct shapes occur with different experiences.

A more accessible example is discussed on page 9 and that is seeing Jennifer Aniston in a movie.  In that case, the complexes at the neuronal level affects the probability of past and future states. Consistent with neuroanatomy many specialized neurons are firing or not firing in the visual system that are associated with Jennifer Aniston as an invariant concept.  Other neurons are associated with other invariant concepts that allow for a fuller description in terms of appearance, age, etc.  All of the elements of the complex are intrinsic information and do not depend on visual inputs for example if dreaming or imagining the actress.

The authors also briefly review some of the experimental evidence that is consistent with the theory. They find that the theory is predictive in number of experimental paradigms. Transcranial magnetic stimulation (TMS) can be applied to to conscious individuals and unconscious (dreamless sleep, general anesthesia) individuals. In the conscious state there is a widespread pattern of activation and deactivation noted with high density EEG.  In the unconscious state cortical response is local or global and stereotypical - integration and information are lost.  A metric called the perturbation complexity index (PCI) a measure of the EEG compressibility from TMS stimulation can be used consciousness and it decreases in states that lack it.   

Tonini has been very explicit about the issue of machine consciousness - it doesn't exist no matter how sophisticated the machine is.  Any machine recognizing inputs that the human nervous system would recognize and producing identical outputs, even if that machine duplicates the structure and function of the human brain - is not conscious.  Tononi uses the consciousness science term zombie to characterize such machines.  By definition a zombie system is one that lacks consciousness and they are described as being subsystems in humans (2) when they are active outside the sphere of conscious recognition.

That brings us back to the ability to detect machines from humans.  If a machine is a perfect human zombie in terms of its input and output, we would not expect an empathy or Turing test to throw it off.   IIT theory acknowledges that what appears to be human input and output can be perfectly simulated.  The original Blade Runner protocol seems more than an empathy test. Specific questions about past memories illustrate an attempt determine if there is continuity between any current and past experiences, even though in the case of Rachael - the memories are false and implanted.

That being said IIT states there there is no Turing test for consciousness.  By now it does seem that fairly basic programs (like self learning neural nets) can replicate a narrowly defined human skill. In that case many people speculate that there is an intelligence or even human consciousness behind it.  On the other hand the perturbation complexity index (PCI) seems like a potentially useful test based on current results.



George Dawson, MD, DFAPA


References:

1: Tononi G, Koch C. Consciousness: here, there and everywhere?  Philos Trans R Soc Lond B Biol Sci. 2015 May 19;370(1668). pii: 20140167. doi: 10.1098/rstb.2014.0167. Review. PubMed PMID: 25823865; PubMed Central PMCID: PMC4387509.

2:  Koch C, Crick F. The zombie within. Nature. 2001 Jun 21;411(6840):893. PubMed
PMID: 11418835.




Addendum:

I read Christof Koch's book Consciousness - Confessions of a Romantic Reductionist a couple of months after this post.  In it he credits Tononi for Integrated Information Theory:

..."The theory of integrated information, developed by the neuroscientist and psychiatrist Giulio Tononi, starts with two basic axioms and proceeds to account for the phenomenal in the world." (p. 6)


     

Monday, July 24, 2017

A New Perspective on Dreaming


From: Reference 1 with permission.


A friend of mine insists on telling me his dream.  He knows I am interested:

"So I am in this old house.  I have the feeling it is my Grandmother's house, but it is really a house I have never been in before.  There is a gathering on the main floor and there are two people there.  I know that one of them is supposed to be my grandmother but it doesn't look at all like her.  There is a guy there who is apparently dating my grandmother.  I know that he is supposed to be a handyman that my brother introduced to my Grandmother but I have never seen him either.  They look like they are in their 70s.  But in reality as you know - my grandmother has been dead for over 20 years."

"Suddenly I am no longer at the party.  I am in the house and I am in an upstairs bedroom.  For some reason, I think it is my grandmother's bedroom.  I don't know why I'm there but all of a sudden this guy comes down the hallway.  He is one of the commentators from TMZ that Hollywood gossip show (let's call him Bob).  I look to the floor next to the wall to my right and there are two small bowls of M&Ms - a bowl of green M&Ms closer to him and a bowl of red M&Ms closer to me.  They are in those Anchor Hocking glass bowls without the blue plastic lids.  Anyway - I grab a small handful and start eating them.  They are dark chocolate M&Ms.  I look at Bob and say: "That's what they're there for" and he starts eating a few."  He asks why I am there and I say: "I heard there was a mouse in here and I need to kill it."

"A mouse runs between me and Bob and I tell him to kill it.  He misses it and it runs at me and I kick at it and knock it into the corner.  Bob is still reluctant to kill it.  The mouse runs at me again but this time it is as big as a rat.  I kick it into the corner of the room again.  This time it runs back out at me and it not longer looks like a rat - it is as big as an otter.  I kick it again - but this time I am shaken awake by my wife.  She asks me if I was having a bad dream.  She said I was kicking my legs like I was running and punching my arms in the air for a few seconds.  My heart was pounding like I was really in a fight with this thing.  What was all of that?  What  does it mean?"

There are a few things about this dream that are striking.  The first is the amount of detail recalled right down to what appear to be the product placements.  Most people telling me about their dreams rarely recall this level of detail.  Often they recall only the emotional tone of the dream and the vague idea that something happened.  The second is the overall content of the dream.  By the dreamer's report it is illogical - none of the events really happened or are likely to happen.  With the exception of the TV celebrity, none of the people in the dream were really who they were supposed to be.  Strangers were supposed to be his grandmother and his grandmother's boyfriend but in reality - there was no such relationship.  There is the movement.  The dreamer is thrashing about the bed until his wife wakes him up.  A final consideration that I like to think about is the processing power necessary to create this experience either de novo or from existing elements.

Dream interpretation is still alive and well in psychiatry - at least the way I practice it.  It is not quite the detailed analysis of all of the elements that Freud thought were important but a combination of a look at the predominate affects and what might be called a synthesis of what is supposed to happen in dreams.  It is also not quite where we need it to be from a neuroscientific perspective.  For example, for the most part we are still operating on a model that suggests more dream activity occurs in REM (rapid eye movement) sleep and that NREM (non-rapid eye movement) sleep contains very little.  We know from dream studies that is not completely correct because both REM and NREM sleep have EEG correlates and we can wake research subjects up during dreams and determine if they are dreaming or not.  Based on those studies there is a rough correlation - but there are still dreams occurring during NREM sleep and REM sleepers without dreams.  Various theories have been advanced about why that occurs, but there is no comprehensive theory.  The other issue is that dream content needs a better explanation.  The simplified explanation is that illogical impossible dreams like the one described here are REM dreams and that NREM dreams are more like plausible events.  Finally - movement during REM dreams is not possible suggesting that the dreamer in the above example was not in REM sleep or he has a neurological problem to account for the dissociation between his motor activity and the fact that he should be paralyzed in REM sleep.  These thoughts about REM and NREM sleep are so pervasive in our society that I routinely interview patients who tell me why they think they are (or are not) getting enough "REM sleep".

I was lucky to have found a recent paper (1) on the subject that if correct may prove to be a landmark study about the neural basis of dreaming and possibly consciousness.  One of the advantages of this paper is that is it written from the perspective of consciousness researchers with an interest in the neural correlates of consciousness.  In this study the authors ran three experiments looking at the question of dream reports and high density (256 channel) EEG.  They used a serial wakening model in which subjects were awakened and asked to report if they were dreaming and could recall some of it (DE = dreaming experience) or if they experienced something but could not recall (DEWR=dreaming without recall of content).  A third option was no experience of dreaming (NE = no experience).  They were asked to characterize any content further according to protocol.  There were two groups of research subjects.  The first was a group of 32 subjects who underwent few awakenings - 233 total.  The second was a smaller group of 7 subjects who had many (815) awakenings.  In a third experiment 7 subjects were studied with 84 awakenings to see if the results of the first two experiments could be predicted.

The initial section of the paper reports on the results of DE versus NE experience in the low frequency (1-4 Hz) power spectrum.  The authors were able to identify what they describe as a posterior cortical hot zone (bilateral parieto-occipital area including the occipital lobe extending to the precuneus and posterior cingulate gyrus superiorly p. 873).  DE occurred when there was decreased low frequency power in this region.  That condition occurred in  both REM and non-REM states.  This finding across distinct sleep stages appears to be highly significant.

The next section of the paper reports on DE versus NE in the high frequency power spectrum (20-50 Hz) that corresponds with high rates of neuronal firing.  Some of the results are summarized in Figure 3 at the top of this post.  In the DE experience condition increased high frequency power was noted in the same parieto-occipital regions that were associated with decreased low frequency power but it was more extensive. DE with recall of content was associated with more widespread extension of the high frequency map than DEWR (no recall of content).  Additional observations were made of the high frequency maps with regard to specific recalled dream content.  The results here are extremely interesting in terms of the specifics of content.  The authors comment on the "perception versus thought" content of dreams.  Some recalled content is an isolated thought or emotion and other content is very vivid imagery including full conversations like the example at the top of this post.  In their experiments, the authors note that there appears to be an anterior -> posterior gradient for high frequency activity with thought content mapping out over frontal cortex and perceptual content mapping out over posterior cortical regions.  They looked at dream content involving facial recognition and noted an increase in high frequency activity over the right fusiform gyrus - a structure noted to be involved in facial recognition during wakefulness.  Dream content that involved spatial imagery was correlated with increased high-frequency activity in the right posterior parietal cortex and area with that expected function during wakefulness.  Additional correlations were noted with movement and speech.

In the final phase of the experiments, the authors sought to find out if the markers identified in the initial sections of the paper could be used to predict where or not a person was dreaming just based on their EEG data.  They were able to accurately predict dreams 80.7 to 91.6% of the time (87% accurate across all states).  

I consider this to be a potentially critical paper to any psychiatrist interested in sleep or dreaming.  If replicated it illustrates that there is a posterior cortical hot zone that correlates with dreaming across REM and NREM sleep stages.  That in itself explains the lack of tight correlation of dreams with REM and NREM sleep.  From a theoretical standpoint they point out the the low delta activity (1-4 Hz) that correlates with dreaming also corresponds to alternations in neuronal depolarization and hyperpolarization that causes a breakdown in cortical communication.  High delta activity  corresponds to states of diminished consciousness including some forms of delirium and loss of consciousness.  They suggest that posterior cortical activation should be studied in patients with disorders of consciousness to see if there may be consciousness without responsiveness based on activity in this area. They also discuss the broader implications of dreaming as a model for the study of consciousness.

That is a good point to end this post.  I will continue to monitor the work of these authors and have been following some of them for some time.  Dr. Tonini for example is probably one of the top experts (and theorists) on consciousness and the only psychiatrist who I am aware of who is doing this work.        


George Dawson, MD, DFAPA


References:

1:  Siclari F, Baird B, Perogamvros L, Bernardi G, LaRocque JJ, Riedner B, Boly M,Postle BR, Tononi G. The neural correlates of dreaming. Nat Neurosci. 2017 Jun;20(6):872-878. doi: 10.1038/nn.4545. Epub 2017 Apr 10. PubMed PMID: 28394322; PubMed Central PMCID: PMC5462120



Attribution:

Figure 3 at the top used with permission from Nature Publishing Group - License Number 4154981341951.  The figure is from reference 1.



Supplementary:

As I have previously posted - I have experience with standard array quantitative EEG (QEEG) and its limitations.  I am a little skeptical of being able to determine the EEG spectrum in the fusiform gyrus by standard surface electrodes in what I imagine is a cap array.  But time will tell.